Is colorectal cancer screening really $1,000 out of pocket? Recommendations, the two stages of charges, four eligibility conditions and six situations that rule you out, 2026

TL;DR The $1,000 covers only the colonoscopy doctor's extra fee. The other two out-of-pocket items — the extra fee at the stool-test stage, and every pre-colonoscopy consultation after the first — carry no official cap at all, so no real ceiling can be written down. As for why not simply have a colonoscopy: the randomised trials concluded "non-inferior", which is not "just as good".

---

#$1,000 is not your whole outlay: three items are out of pocket, and only one is capped

Start with the point most often misread. The $1,000 cap applies only to the colonoscopy doctor's extra fee. It is not the whole of the colonoscopy stage, still less the whole of the programme.

The programme runs in two stages, and the money is structured in two halves as well.

Stage one is a primary care doctor plus the FIT. Each cycle runs 24 months. A negative FIT is subsidised for one consultation, a positive one for two, at $280 each; a first-time participant also brings the doctor a one-off administrative subsidy of $76.

⚠️ But the official page states that doctors are permitted to charge participants a fee on top of the Government subsidy, that the extra fee for the two consultations should be the same amount, and that it is collected at the clinic. That page sets no cap — the Department of Health merely "encourages" primary care doctors to charge no extra fee [Note 1].

The same page lists four things the doctor may not charge extra for: enrolment, the FIT laboratory test, notification of a negative result, and the broadest of the four — a participant contacting the primary care doctor for advice on collecting the specimen, or being reissued two sample bottles where needed. Which is to say that ringing back to ask how the collection works cannot attract an extra fee either, not only the reissue of bottles.

Stage two is the colonoscopy doctor. The official page states that the extra fee should be no more than $1,000, charged once, and set out on joining the programme. The Department of Health's press release of 31 December 2025 states that over 70% of service locations charge no extra fee for colonoscopy or polyp removal.

⚠️ But the paragraph above it on the same page describes another out-of-pocket item with no cap, and it is easy to miss. The Government subsidises one pre-colonoscopy consultation per participant ($300), and pays it only to the first colonoscopy doctor to submit a subsidy claim; the official wording is that participants have to pay for subsequent pre-colonoscopy consultations themselves [Note 2].

Put plainly: if you see a second colonoscopy doctor — to compare prices, to get a second opinion, or because the first will not do it — that consultation is entirely at your own cost, and the Government has set no cap on it.

Three cases worked with the figures already quoted, over one 24-month cycle:

• A: FIT negative, doctor charges nothing. You pay $0; the Government spends $280 + $76 = $356.
• B: FIT positive, colonoscopy done with a polyp removed, neither doctor charges extra. You pay $0; the Government spends $280 × 2 + $76 + $8,500 = $9,136.
• C: as B, but both doctors charge everything they may. You pay (the primary care doctor's extra fee × 2) + $1,000 + (the cost of each pre-colonoscopy consultation after the first × the number of them).

Of the three, only the $1,000 in the middle has an official cap; the first and last have none, so the total is a sum official figures cannot answer and only the doctors you book with can.

And one item is addressed specifically to people in financial hardship. Immediately after the "no more than $1,000" sentence, the same page states that the Department of Health encourages colonoscopy doctors who charge an extra fee to offer charity quotas free of extra fee to recipients of Comprehensive Social Security Assistance and holders of a Certificate for Waiver of Medical Charges [Note 3].

⚠️ That is encouragement, not a requirement, and you have to ask.

---

#Five questions worth asking before you book

1. Do you charge an extra fee, and how much?
2. If the FIT is positive and a second consultation is needed, is it the same amount again?
3. What extra fee does the colonoscopy doctor you refer to charge — is it $0?
4. If I switch to a second colonoscopy doctor, what will that pre-colonoscopy consultation cost me?
5. If I receive CSSA or hold a Certificate for Waiver of Medical Charges, do you have a charity quota?

---

#All four conditions must hold, and six situations rule you out

All four must hold. If any one of them does not, the answer is to see a doctor about a different route, not to assume nothing is wrong.

The programme's eligibility page lists four [Note 4]:

The eligible birth years for 2026 are 1950 to 1976.

The situations in which a person should not join, as the official page lists them, are six: symptoms of colorectal cancer; a past colorectal cancer, or an immediate family member who has had one; a history of chronic inflammation of the colon such as ulcerative colitis; a confirmed hereditary bowel disease, or an immediate family member with one; a history of colorectal polyps; and being confirmed by a doctor as unsuitable for colonoscopy.

⚠️ That page does not say these six exhaust the possibilities, and the last of them is itself a catch-all handed back to the doctor. The same page states that the participating primary care doctor will first ask about the participant's condition and relevant medical history and complete enrolment only after ruling the above out. So "I am not in any of those six" does not mean "I am certainly suitable".

Three situations mean you need not join: a colonoscopy within the past ten years; a sigmoidoscopy within the past five years; a faecal occult blood test within the past two years.

⚠️ The same page explains why these three can substitute for one another, but that explanation opens with a condition that cannot be cut away: it is speaking of those "without a significant family history". The sentence that the three methods are of comparable effect is addressed to people without a significant family history; it does not apply to those with one, for whom the advice is different [Note 5].

Collection uses two bottles, one sample from each of two separate bowel movements, returned within 4 days counting from the day of the first collection — the official wording is "in general", not a hard boundary. Two further points: too much or too little sample will both affect the result; and before returning it the specimen should be kept somewhere cool.

What happens afterwards? Negative: repeat in two years. Positive: referral to a colonoscopy doctor in the programme. Polyps found: removed during the same examination. Colorectal cancer confirmed: treatment is for the doctor to decide.

---

#Why birth year rather than age

1950 + 76 = 2026, so someone born in 1950 turns 76 during 2026; 1976 + 50 = 2026, so someone born in 1976 only turns 50 during 2026.

Within a single year, then, the list takes in both people who have not yet had their 50th birthday and people who are already past their 76th. "Aged 50 to 75" is a label; day by day it does not line up. The press release also states that the eligible birth years are updated at the start of each year.

---

#Symptoms take the consulting route, not the screening route

The Government's Cancer Online Resource Hub page on colorectal cancer opens with it: early colorectal cancer may have no symptoms at all.

The lead-in to that page's symptoms section reads "common symptoms of colorectal cancer include:" — "include" is the source's own word, so the five items below are the common symptoms that page lists, not an exhaustive list:

• blood in the stool, or stool that is black or contains mucus, or rectal bleeding;
• a sudden change in bowel habit (persistent constipation or diarrhoea), a change in the shape of the stool (narrow and thin), or a feeling of incomplete emptying after passing stool;
• unexplained weight loss;
• lower abdominal discomfort (bloating or cramping);
• signs of anaemia: cold hands and feet, tiredness, a faster heartbeat, shortness of breath, pallor, dizziness.

⚠️ A symptom outside these five does not mean nothing is wrong, and it does not mean you may treat yourself as "without symptoms" and join the screening programme.

The next section of the same page states that people at high risk and people with the common symptoms of colorectal cancer should consult a doctor and be examined regularly or as soon as possible. Symptoms take the consulting route, not the screening route.

---

#The family-history tier has a threshold, and the starting age is not only 40

If someone in your family has been diagnosed with colorectal cancer, that immediate family history is exactly what puts you outside this programme. But the "different advice" has to be given in full, because it has a threshold and its starting age follows two rules [Note 6].

First, this advice is not "applies if a relative has had it". The condition is one immediate family member diagnosed at 60 or below, or more than one immediate family member diagnosed at any age. A reader whose parent was diagnosed at 72 is not in this tier — still not eligible for the programme, but also not entitled to assume that arrangement applies.

Second, the starting age is not simply 40. The source reads: from age 40, or 10 years before the age at which the youngest affected relative was diagnosed, whichever is earlier (but not before age 12). If a sibling was diagnosed at 45, the start is 35, not 40.

The same page also sets out the route for hereditary bowel disease, and these are precisely the people excluded by the "should not join" list above: people carrying a Lynch syndrome mutation should have a colonoscopy every 1 to 2 years from age 25; people carrying a familial adenomatous polyposis mutation should have a sigmoidoscopy every 2 years from age 12.

⚠️ All of the above is for a private conversation with a doctor, not for matching yourself to a row.

---

#The FIT tests for blood, not for cancer: 0.74 against 0.40

The FIT tests for blood in the stool that the eye cannot see, not for cancer cells; a positive means only "there is blood, and it has to be looked into".

Three pairs need keeping apart first. Screening is not diagnosis — the programme's own definition carries two conditions, not one: it is for people who (1) have no symptoms and (2) are suitable for colonoscopy. A positive is not a diagnosis — it means only that blood was detected, and where the blood comes from is for the colonoscopy to establish. A polyp is not a cancer — most of what is removed are adenomas, that is, polyps that can turn into cancer.

The FIT's own figures are in the document the programme cites.

⚠️ In one sentence: a single FIT catches roughly seven of every ten colorectal cancers, but only four of every ten advanced adenomas.

And that paper states the limitation plainly: these are the figures for a single test, not for testing repeated over time [Note 7] — and repeating every two years is exactly how the programme is designed. That is why repetition is the core of the mechanism, and it is the quantitative basis for "a negative does not mean nothing is wrong".

As for why most positives are not cancer, the mechanism is written in the WHO's guide: as prevalence falls, the positive predictive value falls with it, without sensitivity or specificity changing at all [Note 8].

---

#One of the programme's published percentages comes without a denominator

The programme's own published results have to be read together with their denominators — and one of the press releases does not give one.

The Department of Health's press release of 31 December 2025 states: as at the end of November 2025, more than 66 000 people had newly joined during the year and total participation exceeded 579 000; 15% of faecal immunochemical test results were positive, of whom about 45 000 (7.9%) were diagnosed with colorectal adenoma after colonoscopy and about 3 800 (0.7%) with colorectal cancer.

That release does not state which group those bracketed percentages are taken against, and they do not all reconcile to the total of 579,000.

Take all participants as the denominator: 3,800 ÷ 579,000 = 0.66%, which rounds to 0.7% and agrees; but 45,000 ÷ 579,000 = 7.77%, not 7.9%. Take those who tested positive instead: 579,000 × 15% ≈ 86,850 people, and 45,000 ÷ 86,850 ≈ 51.8%, which is not 7.9% either.

Neither reading reconciles, so this article reproduces the release's figures as published, does not assign a denominator on its behalf, and does not divide these percentages by one another to build ratios.

In another document the programme uses stratified denominators, and says so plainly. The Centre for Health Protection's Non-Communicable Diseases Watch, against a background of over 510,000 participants as at the end of 2024, states that among participants who returned a stool sample and obtained an analysable result, about 77 000 (about 16%) had a positive immunochemical test result; and that among the positive cases who underwent colonoscopy, the programme had found about 3 400 colorectal cancers and about 40 000 colorectal adenomas [Note 9].

Neither denominator is all participants. Of about 77 000 positive results, the colonoscopies eventually found about 3 400 colorectal cancers: 3 400 ÷ 77 000 = about 4.4%.

⚠️ This article did that division itself; the programme has not published this ratio, and the numerator and denominator do not describe quite the same group — the denominator is everyone who tested positive, while the numerator counts only those who actually went for colonoscopy and had a cancer found at it. Among positives who never went, there may be cases this article has not counted, so the figure can only be read as a rough lower bound: the true positive predictive value can only equal or exceed 4.4%.

Even taken as a lower bound the number is small: the commonest outcome of a positive is not cancer but "a polyp was removed" or "nothing was found".

---

#Why not simply have a colonoscopy? "Non-inferior" is not "just as good"

Because at the level of a whole programme, a test of moderate sensitivity that more people are willing to do and that repeats every two years can reach non-inferiority — "not worse than", not "just as good".

The backbone is the 10-year result of Spain's COLONPREV, still the only randomised trial to set 10-year colorectal cancer mortality as its primary endpoint and put the two programmes side by side. It enrolled people aged 50 to 69 with no personal or family history of colorectal cancer:

The risk difference was −0.02 (95% CI −0.10 to 0.06) and the RR 0.92 (95% CI 0.64 to 1.32), with non-inferiority p=0.0005 against a margin of 0.16 percentage points.

⚠️ The authors' conclusion carries a condition that cannot be cut away: non-inferiority holds "based on the participation rates observed in this study", which makes it a programme-level conclusion, not a test-level one [Note 10]. And the RR's bounds (0.64–1.32) state plainly that the result does not exclude colonoscopy being some 30% better on mortality.

The mechanism is here, and it needs all three inputs.

One, the trial design. What is compared is a one-off colonoscopy against a FIT every two years — one chance for the colonoscopy arm, several rounds within ten years for the FIT arm.

Two, the first-round programme-level yield. In the same trial's first round: advanced adenomas in 514 people (1.9%, colonoscopy arm) against 231 (0.9%, FIT arm); non-advanced adenomas in 1,109 (4.2%) against 119 (0.4%); but almost the same number of cancers, 30 (0.1%) against 33 (0.1%). The denominators of those percentages are everyone randomised, not everyone examined — in the first round only 24.6% of the colonoscopy arm and 34.2% of the FIT arm were actually examined. Which is to say this set measures "how much a programme finds after one round", not "how capable a single examination is".

Three, participation. Counted as "had any screening at all", the FIT arm reached 39.9% against the colonoscopy arm's 31.8% (RR 0.79, 95% CI 0.77–0.82), a factor of 1.25. But that 31.8% is not the proportion who had a colonoscopy. Broken down: in the FIT arm 39.4% did the FIT and 0.5% went for a colonoscopy; in the colonoscopy arm 20.1% had the colonoscopy and 11.7% did a FIT instead. The proportion who did the test they were assigned is 39.4% against 20.1%, a factor of 1.96 — not 1.25.

Work the yields the same way: advanced adenomas 514/26,703 = 1.92% against 231/26,599 = 0.87%, a factor of 2.22; non-advanced adenomas 1,109/26,703 = 4.15% against 119/26,599 = 0.45%, a factor of 9.28 [Note 11].

A 1.96-fold difference in willingness to do the assigned test still cannot arithmetically offset a 2.22-fold to 9.28-fold difference in yield; what makes up the rest is repetition. "More people willing" plus "again every two years" is what together brings ten-year mortality to non-inferiority.

Two further trials point in a not entirely identical direction, and this article does not skip them.

NordICC is the largest and longest-followed randomised trial of inviting people to colonoscopy against not inviting them. At 10 years: 28,220 in the invited group and 56,365 in usual care, but only 11,843 of the invited (42.0%) actually went; intention-to-screen colorectal cancer risk 0.98% against 1.20% (RR 0.82, 95% CI 0.70–0.93), death from colorectal cancer 0.28% against 0.31% (RR 0.90, 95% CI 0.64–1.16), all-cause mortality 11.03% against 11.04%; 15 people had major bleeding after polyp removal, with no perforation or screening-related death within 30 days.

The 13-year follow-up analysed 84,583 people: colorectal cancer incidence 1.46% (375 cases / 28,217 people) against 1.80% (912 cases / 56,366), intention-to-screen RR 0.81 (95% CI 0.71–0.90); colorectal cancer death 0.41% (106) against 0.47% (236), intention-to-screen RR 0.88 (95% CI 0.68–1.08). The authors conclude that a single colonoscopy significantly reduced incidence at 13 years, while mortality had still not reached significance [Note 12].

The same 13-year report also answers the question of the person who will actually go. The per-protocol estimates: colorectal cancer incidence RR 0.55 (95% CI 0.33–0.81) and colorectal cancer death RR 0.70 (95% CI 0.26–1.25) — incidence significantly lower among those who really were examined, but the confidence interval for mortality still crossing 1.

Sweden's SCREESCO randomised 278,280 people aged 60 into three arms: a one-off colonoscopy, two rounds of two-sample FIT (10 µg/g threshold), and a usual-care control with no screening. At a median follow-up of 4.8 years, the diagnostic-phase results: colorectal cancer incidence 107.9 per 100,000 person-years in the colonoscopy arm against 99.9 in the control (IRR 1.08, 95% CI 0.91–1.28), and 96.0 against 103.9 in the FIT arm (IRR 0.92, 95% CI 0.81–1.05) — neither arm significantly better than the control; though detection of stage I or II cancers was higher in both than in the control (colonoscopy IRR 1.38, 95% CI 1.09–1.74; FIT IRR 1.19, 95% CI 0.99–1.43).

The authors state that these are diagnostic-phase results and that the primary endpoint is incidence rather than 10-year mortality, so it is not a counter-example to COLONPREV's claim of uniqueness, but it is the most direct evidence in the other direction this field has.

For balance: a post-hoc COLONPREV analysis adjusted for adherence (17,270 people who began screening) found that if both strategies are kept up, 10-year colorectal cancer mortality is lower with colonoscopy (FIT 0.072% against 0.042%, risk difference −0.03%, 95% CI −0.12% to 0.06%), but the authors state the estimate is imprecise, and incidence and all-cause mortality were comparable between the arms.

So the accurate statement is not "FIT is as good as colonoscopy" but "at programme level, a FIT-led programme is not worse than a colonoscopy-led one".

As for the three older randomised trials of guaiac faecal occult blood testing, they point the same way but three things have to be said with them [Note 13]: Minnesota's 33% came from rehydrated slides in a volunteer population; the Funen trial did not measure a fall in colorectal cancer incidence (481 cases in the screened group against 483 in the control, the reduction being in deaths only); and all three used the guaiac method, not today's immunochemical one.

---

#None of the three randomised trials covers ages 70 to 75

⚠️ This boundary has to be stated.

COLONPREV enrolled 50 to 69, NordICC 55 to 64, and SCREESCO everyone at 60. The Hong Kong programme runs to 75, which means that for the band from 70 to 75 none of the mortality or yield figures above has direct randomised support; it is extrapolated from the expert group's judgement and overseas practice.

On the age boundary itself, the official reasoning is worth noting — it is in the expert group's own document, and it runs to five points, not one sentence. The document first sets out the assessment study the Government commissioned from the University of Hong Kong, whose simulation model suggested that lowering the target range to 40–70 might be more favourable in life-years and QALYs; the strongest argument against the present boundary is one the group itself cites. It then lists five considerations the other way: a higher cost and a higher number of colonoscopies per colorectal cancer or per death averted; no significant rise in recent years in colorectal cancer incidence or share among those aged 40–49; a low proportion of people aged 50 to 75 ever screened for colorectal cancer (40%); rising demand for surveillance colonoscopy generated by adenomas found; and most overseas health authorities not listing 40–49 as a recommended target.

The group's conclusion is that rather than widening the age range, resources are better spent raising participation and adherence in the existing target population. Another sentence from the same document belongs here too: moving from 50–75 to 40–70 could miss the 20% of colorectal cancers the programme currently finds in people aged 71 to 75.

One more distinction: "aged 50 to 75" and "a FIT every two years" are two different lines. What the expert group sets out for people at average risk is a choice of three methods at a choice of frequencies — faecal occult blood testing every year or every two years, sigmoidoscopy every five years, or colonoscopy every ten years [Note 14]; the Government's programme (piloted in 2016, made permanent in 2018, extended fully in 2020) runs one frequency of one of those methods.

---

#Screening cannot do three things, and the follow-up carries risk of its own

Three things it cannot do: it will not be completely accurate, it will generate false alarms that must be followed up, and the follow-up itself carries risk.

The programme states that the test will not be completely accurate, and states the consequence of a false negative too: someone who is in fact ill may be wrongly reassured by a negative and delay seeking care or treatment, "thus affecting the prognosis" [Note 15].

⚠️ Symptoms appearing after a negative cannot be soothed away with "I was only just tested". And a single FIT's sensitivity for advanced adenoma is only about 0.40, so a negative has limited power to rule polyps out.

The WHO's guide explains why the weight of harm is greater than it seems: because most of those screened do not have the disease, more people can be exposed to the harm of screening than may be able to benefit from it. The same guide notes that every positive result requires further examination, so those with a false positive bear unnecessary tests and complications — one of its examples being bowel perforation after colonoscopy.

How high is the complication rate for colonoscopy? Both sets of figures are given, because they measure different things:

The death figure for the first set is 2.9/100,000 (95% CI 1.1–5.5, about 0.03/1,000). And that pooled paper states it itself: complication rates for screening and surveillance are lower than for diagnostic examinations.

So the all-indications figures above run about 1.6 to 1.8 times higher (0.5 ÷ 0.31 = 1.6; 2.6 ÷ 1.46 = 1.8). The second set is the order of magnitude you face when you join a screening programme.

The Hong Kong Government has a qualitative statement of its own: more serious complications such as bowel perforation, bleeding, peritonitis, cardiopulmonary complications, infection and acute bowel obstruction rarely occur, and in general the chance of a serious complication is less than 1%. But if abdominal pain or blood in the stool appears after the examination, seek care immediately [Note 16] — that last sentence is the safety net after a colonoscopy.

A related point: the programme's statement that over 50% of the cancers it confirms are early stage is not in itself evidence that screening works — lead-time bias and overdiagnosis both make screen-detected cases look earlier [Note 17]. The real evidence is the randomised mortality figures in the section above.

In fairness, though: the figure behind that "just over 50%" and the definition of "early" are both obtainable. The definition is early = stage II or below: about 56% of the colorectal cancers confirmed through the screening programme are early stage (a Legislative Council reply uses the same definition and gives about 57%). The territory-wide comparison is the 2023 stage distribution: stage I 9.8%, stage II 19.4%, stage III 26.6%, stage IV 23.9% and unstaged 20.4% — so stage I and stage II together come to 29.2% territory-wide.

⚠️ The contrast is real, but it is still not mortality evidence — the two biases above are precisely what produce a stage gap like this; and 20% of the territory-wide set (20.4%) is "unstaged", so that denominator is not entirely clean.

As for why stage matters so much, the Government's Cancer Online Resource Hub sets out five-year relative survival by stage — stage I 95.7%, stage II 87.3%, stage III 68.7%, stage IV 9.3% — and gives 58.2% for patients overall on the same page.

Two things need keeping apart: the Cancer Registry's chart labels those four by-stage figures as cases diagnosed from 2010 to 2017; but the 58.2% is not among the figures carrying that label — the overall five-year relative survival marked 2010–2017 in the same statistics is 57.3% for men and 59.6% for women. So this article reports 58.2% only as the overall figure written on the Cancer Online Resource Hub page, and does not attach the 2010–2017 label to it.

---

#The Hong Kong figures: case numbers up, rates down, younger adults up

The total number of cases rose 14.6% over ten years, but the age-standardised incidence rate fell over the same period — and it fell at 50–74 while rising at 25–49. All three sentences are true and they do not contradict one another.

The 2023 baseline: 5,467 new cases, 14.4% of all new cancers, third by rank; 2,266 deaths, 15.2% of cancer deaths, second by rank.

ASR is the age-standardised rate, computed against the Segi 1960 world standard population.

The Registry's own ten-year trend indicator has flipped. The average annual percentage change in ASR over the past ten years (male/female): to 2019, +0.6%/−0.3%; to 2021, +0.4%/−0.4%; to 2023, −1.8%/−1.7%. All four are statistically significant at the 0.05 level.

Over the same period the total case count rose: new colorectal cancers went from 4,769 in 2013 to 5,467 in 2023, a rise of 14.6% (over the same period all cancers went from 28,936 to 37,953, a rise of 31.2%).

How can case numbers rise while the rate falls? The official overview answers it at paragraph 1.10, and the sentence gives three causes, not one: driven mainly by population ageing, together with changes in cancer risk itself and improvements in diagnosis [Note 18]. Another passage adds that new cancers of all kinds grew 2.7% a year over the past decade while the total population grew only 0.5% a year and the population aged 65 and over grew 4.9% a year. Age standardisation exists precisely to offset that effect, which is why case numbers and rates always have to be read together.

There is a split between age groups too. The official key findings state that colorectal cancer is rising in younger adults while the rates in older adults of both sexes have fallen since 2017. Average annual percentage change 2014–2023, with 2023 case numbers: men aged 50–74, −1.6% (2,052 cases) and 25–49, +1.5% (164 cases); women aged 50–74, −0.6% (1,332 cases) and 25–49, +0.6% (171 cases).

⚠️ But none of that supports saying "the screening programme made it fall". This is coincidence in time, not proof of cause; the overall trend line is simultaneously affected by better treatment, changes in diagnosis, and shifts in diet and smoking rates. The WHO's guide lists exactly this trap: comparing mortality figures before and after a screening programme is introduced can mislead, because the improvement may come from better population-wide diagnosis and treatment rather than from the programme. Attribution needs randomisation and a side-by-side comparison. The Registry also cautions that a trend is ideally observed over five years or more before it can be read.

---

#Australia posts the kit to you — but at 45 to 49 you have to ask for it too

Same test, same disease, and the two defaults are broadly opposite: Australia posts to everyone aged 50 to 74, while Hong Kong waits for everyone to enrol. But for Australians aged 45 to 49, Australia waits for you to ask as well.

The Australian programme uses an immunochemical faecal occult blood test, and eligibility has three conditions, not one: aged 45 to 74; holding a green Medicare card or registered as a Department of Veterans' Affairs client; and having an Australian postal address. The Medicare enrolment file is the list the kits are posted from.

And "posted to you automatically" has to be said age band by age band. The official page says it itself: people aged 50 to 74 receive a kit every two years; people aged 45 to 49 can request their first kit through an online form or by telephone [Note 19].

Which is to say that at 45 to 49, Australia waits for you to ask — the same default as Hong Kong's. In Hong Kong you first enrol in eHealth, then arrange to see a participating primary care doctor yourself, and the doctor issues the sample bottles after the consultation.

How many Australians actually do it? The figure exists, but it has to be read with its denominator. Of the 5,744,244 eligible people invited between 1 January 2018 and 31 December 2019, 2,497,317 took part, a participation rate of 43.5%; the preceding rolling two-year period (2017–2018) was 42.4%.

⚠️ Three qualifications belong with it: the denominator is those invited, not the whole population; the target age range for that reporting period was 50 to 74, with 45 to 49 added only on 1 July 2024; and participation is measured over rolling two-year periods, so any two consecutive published periods overlap by a calendar year and 43.5% against 42.4% cannot be read as "up 1.1 percentage points year on year".

What that structural difference means is not for us to assert. The WHO's guide splits the "how the test is offered" dimension in two: ad hoc, relying on individuals to take up the offer or refer themselves, with the consequence that participation tends to be low and those who use the service tend to be of higher socioeconomic status; and call and recall from a register of the eligible population, with the consequence of raising participation and narrowing socioeconomic inequalities [Note 20].

The two territories' figures cannot be subtracted. Australia has an invitation — the posted kit — so what it measures is whether the invited respond; Hong Kong has no invitation mechanism and therefore no corresponding denominator at all. The accurate statement is not "Hong Kong has only one measure" but that Hong Kong has no measure that lines up with Australia's.

What Hong Kong most often cites is a population survey: the proportion of people aged 50–75 who have "ever" had colorectal cancer screening rose from 33.2% in 2014/15 to 42.6% in 2020–22 and 46.2% in 2023, and the same document states plainly that the screening rate still needs improvement.

⚠️ Those three points come from two different survey instruments — 33.2% and 42.6% from the Population Health Survey, 46.2% from the 2023 Behavioural Risk Factor Survey — so they are not a clean time series and should not be read as three consecutive points on a rising line. (The expert group's 2024 document cites a different "ever screened" proportion of 40%, from another source, and this article does not merge the two.)

Asking "have you ever had one in your life" is already a far looser question than "did you respond to an invitation within two years", so even that 46.2% only supports saying Hong Kong is nowhere near universal coverage.

---

#What to do next

1. Check the four eligibility conditions first. All four must hold; if any one does not, see a doctor about a different route rather than assuming nothing is wrong.
2. Then check the six situations that rule you out. Not being in those six still does not mean you are suitable — the last gate is the primary care doctor's assessment at the consultation.
3. Before booking, ask the five questions above. Of the three out-of-pocket items, only the $1,000 has an official cap.
4. If you receive CSSA or hold a Certificate for Waiver of Medical Charges, ask about a charity quota. It is encouraged, not required, and it will not be offered unless you ask.
5. With symptoms, see a doctor; do not take the screening route. The five the Government lists are the common symptoms, not an exhaustive list.
6. If blood in the stool or a change in bowel habit appears within the two years after a negative, go straight to a doctor. A single FIT's sensitivity for advanced adenoma is only about 0.40.
7. If you are in the family-history tier, check the threshold yourself. One immediate family member diagnosed at 60 or below, or more than one at any age; the starting age may be earlier than 40.
8. If abdominal pain or blood in the stool follows a colonoscopy, seek care immediately. That is the safety net the Government sets out.

---

#Frequently asked questions

Is $1,000 everything I have to pay?

No. The $1,000 is only the cap on the colonoscopy doctor's extra fee. Two other items have no cap: at the stool-test stage the doctor may charge on top of the $280 subsidy; and the Government subsidises one pre-colonoscopy consultation only, with every one after that at your own cost. Ask directly before booking. Recipients of CSSA and holders of a Certificate for Waiver of Medical Charges can ask the doctor whether a charity quota is available.

Does a positive FIT mean cancer?

No. A positive means blood the eye cannot see was found in the stool, and where it comes from is for the colonoscopy to establish. The programme's own stratified figures: as at the end of 2024 there were about 77 000 positive results, and among positive cases who went on to colonoscopy about 3 400 colorectal cancers were found — the great majority of those who test positive are not diagnosed with colorectal cancer.

Why a stool test rather than going straight to colonoscopy?

Because at programme level a FIT-led programme reaches "non-inferiority" — not worse than, which is not the same as just as good, and the conclusion carries a condition: it holds on the basis of the participation rates observed in that trial. COLONPREV's 10-year risk of death from colorectal cancer: colonoscopy arm 0.22% (55 deaths), FIT arm 0.24% (60 deaths); the proportion who did the test they were assigned was 39.4% in the FIT arm against 20.1% in the colonoscopy arm, a factor of 1.96. That participation gap alone does not offset the first-round yield gap (advanced adenomas a factor of 2.22, non-advanced 9.28); it has to be read together with the design, a one-off colonoscopy against a FIT every two years, where repetition makes up the rest. If both are kept up, a post-hoc analysis shows lower mortality with colonoscopy, but the estimate is imprecise.

Why are this year's birth years 1950 to 1976?

Eligibility is defined by birth year and updated at the start of each year. 1976 + 50 = 2026 and 1950 + 76 = 2026, so age counted day by day does not line up with the label "aged 50 to 75".

Can I stop worrying for two years after a negative?

A negative is generally followed by a repeat in two years. But the programme states that the test will not be completely accurate and that a false negative can wrongly reassure someone into delaying care, thus affecting the prognosis — so symptoms appearing in between call for a doctor. In numbers: a single FIT's sensitivity is about 0.74 for colorectal cancer and only about 0.40 for advanced adenoma.

I am in my 40s and I keep seeing news about colorectal cancer rising in younger people. Should I be screened now?

The rise is real (+1.5% a year in men aged 25–49), but the 2023 case counts in that band were 164 in men and 171 in women, on a small base. Symptoms or a family history are a different matter and call for a doctor directly. The expert group has also set out five considerations against including 40–49, and chose to concentrate resources on raising participation in the existing target population.

The Government has not written to me. Does that mean I do not need this?

No. Hong Kong has no notification mechanism at all, and not receiving a letter says nothing about your eligibility. Australians aged 45 to 49 likewise have to ask, so "did a letter arrive" has never been the test of whether you need screening.

Will this article give prices for a self-paid colonoscopy outside the programme?

No. This article makes no price comparisons; this platform has a separate article on gastroscopy and colonoscopy charges.

---

#What this article does not state

• The amount or distribution of primary care doctors' extra fees. The official page only encourages charging nothing; it sets no cap and publishes no distribution of charges, so this article puts no figure on this item.
• Other causes of a positive FIT, such as haemorrhoids. The official page lists no such causes, so this article says only "there is blood, and it has to be looked into".
• Complication statistics for the Hong Kong programme's own colonoscopies. No complication figures published by the Colorectal Cancer Screening Programme itself, with denominators, could be found; this is "could not find" rather than "confirmed not to exist" — the full text of the programme evaluation study commissioned from the University of Hong Kong by the Food and Health Bureau and completed in August 2023 could not be obtained, and that report is the document most likely to carry such figures. The complication rates quoted here are international pooled data, and their scope is stated in the text.
• The extent of overdiagnosis in Hong Kong. There is no local quantitative estimate; this article gives only the definition and the mechanism. The WHO's definition reads: "Overdiagnosis identifies a condition or problem that would never cause a person harm during their lifetime".
• The denominators of the 15%, 7.9% and 0.7% in the Department of Health's press release of 31 December 2025. That release does not state which group those percentages are taken against; against total participation of 579,000 the 0.7% agrees (0.66%) while the 7.9% does not (7.77%), and against the number testing positive it does not agree either (about 51.8%). This article therefore reproduces the release's figures, assigns no denominator on its behalf, and does not divide these percentages by one another to derive ratios.
• Why the percentages and case counts in NordICC's 13-year report do not reconcile. The paper's own 1.46%/1.80%/0.41%/0.47% are not "cases ÷ people" (375 ÷ 28,217 = 1.33%, 912 ÷ 56,366 = 1.62%, 106 ÷ 28,217 = 0.38%, 236 ÷ 56,366 = 0.42%), so the case counts should not be divided out to check them. The abstract gives no reason, the full text is behind a paywall, and this article could not verify it; the inference that these are cumulative risk estimates is this article's, not something the paper states.
• The latest figures from the current version of Australia's monitoring report. The Australian participation rates given here are taken from a web archive capture of that report (reporting period 2018–2019). Measured on 9 August 2026: aihw.gov.au returned HTTP 403 (a meaningless path tested in the same window also returned 403, so this is a site-wide block on automated access rather than missing data); health.gov.au and ncci.canceraustralia.gov.au both failed to respond (HTTP 000); the control, colonscreen.gov.hk, returned 200.
• The "last updated" date the Australian programme page states for itself. Every description of the Australian system quoted here is taken from a web archive capture of that page dated 2026-02-18, as stated in the text; this article does not give the page's self-reported last-updated date and does not claim the current version matches the archived one.
• A direct comparison of Hong Kong and Australian participation rates. The denominators differ. This article gives each territory's figures separately but does not subtract, divide or juxtapose them as a comparison of higher and lower. That holds regardless of whether the Australian figures can be retrieved.
• Colonoscopy waiting times in public hospitals. The Hospital Authority has stated in a Legislative Council reply that it does not keep this statistic.
• Prices for a self-paid colonoscopy outside the programme. This article makes no price comparisons.
• This article assesses no doctor, clinic or organisation, and gives no medical advice on any individual situation.

---

#Notes: the official wording

Where a source publishes in English, the wording below is the source's own. Where a source has no English edition, the rendering is ours and is marked as such, with the original set out alongside so it can be checked.

[Note 1] The primary care doctor's extra fee, from the official English page:

The enrolled PCD is allowed to charge the participant a co-payment on top of government subsidy for the subsidised consultation. The amount shall be the same for the first and second consultations, payable at enrolled PCD clinic. However, PCD is encouraged to adopt a zero co-payment amount, to promote participation of the screening service.

The Chinese edition of the same page is more specific about who is doing the encouraging. The English "is encouraged" does not say by whom; the Chinese states that it is the Department of Health that so recommends. Its wording:

扣除政府資助後，基層醫療醫生可就獲資助的的診症服務收取額外費用。第一次及第二次診症的額外費用金額應該相同，並於診所中收取。然而，為鼓勵更多巿民能夠參加篩查服務，衞生署建議基層醫療醫生不收取額外費用。

The duplicated character in 「獲資助的的診症服務」 is the source's own typographical slip, reproduced here as it stands.

[Note 2] The subsidy for the pre-colonoscopy consultation:

A first subsidy of $300 will be paid for pre-procedural consultation. To promote continuity of care, the first subsidy for each participant will only be paid to the first CS who submits claim for subsidy in CRC IT System (i.e. the first CS who records the pre-procedural consultation of the participant in the CRC IT System). All subsequent pre-procedural consultation(s) will be paid out-of-pocket by the participant.

[Note 3] Charity quotas. The Chinese edition runs the point across two sentences with a stray punctuation mark between them; the English edition puts it in one:

If co-payment amounts are set, enrolled CS are encouraged to offer charity quotas by waiving this amount for individuals who are recipients of Comprehensive Social Security Assistance or holders of Certificate for Waiver of Medical Charges.

The sentence before it in the English recommends that the co-payment be set at zero to promote participation by the less well-off members of society; the Chinese edition attributes that recommendation to the Department of Health.

[Note 4] The four eligibility conditions, reproduced from the English page with its own note markers: "Aged between 50 and 75"; "Hold a Hong Kong Identity Card# or Certificate of Exemption"; "Registered in the Electronic Health System (eHealth) (please visit www.ehealth.gov.hk for details)"; "Do not have symptoms of colorectal cancer, higher colorectal cancer risk, or screening/examination for colorectal cancer within a specified period* (To learn more, please click here)". Its two footnotes: "Except those who obtained their Hong Kong Identity Card by virtue of a previous permission to land or remain in Hong Kong granted to them and such permission has expired or ceased to be valid"; "The primary care doctors will personally meet the prospective participants so as to assess their medical fitness for joining the Programme". The Chinese edition is equivalent in content but differs in wording — its fourth item opens with 「但」, which the English does not — and the two pages carry different revision dates, 2025-01-02 for the English against 2025-04-07 for the Chinese. Each is reproduced from its own edition and the two are not merged. The programme's public pages write "FOBT" while its service pages and press releases write "FIT"; what is actually done is a FIT.

[Note 5] The condition under which the three methods substitute for one another:

Local experts recommend individuals who do not have significant family history can consider to receive (1) faecal occult blood test every one or two years or (2) sigmoidoscopy every 5 years or (3) colonoscopy every 10 years, for colorectal cancer screening.

Since the effectiveness of the above three methods are similar, it is not necessary to enroll in the Programme to repeat the screening within the recommended screening interval or with different screening methods.

"Of similar effectiveness" is said of people without a significant family history.

[Note 6] The advice for family history and hereditary bowel disease, from the Government's Cancer Online Resource Hub:

For individuals with one first-degree relative diagnosed with CRC at or below 60 years of age, or more than one first-degree relatives with CRC irrespective of age at diagnosis, colonoscopy should be performed every 5 years beginning at the age of 40 or ten years prior to the age at diagnosis of the youngest affected relative, but not earlier than 12 years of age. As an alternative, the concerned individuals may consider FIT every one or two years after understanding the pros and cons of FIT as compared with colonoscopy.

The same page also states: for carriers of a mutated Lynch syndrome gene, screening by colonoscopy every 1 to 2 years from age 25 onwards; and for carriers of a familial adenomatous polyposis mutation, sigmoidoscopy every 2 years from age 12.

[Note 7] FIT accuracy. Paragraph 14 of the Expert Working Group's professional document of June 2024:

Regarding the accuracy of FIT for CRC detection using threshold of 20mcg haemoglobin per gram of stool, the pooled sensitivity was 0.74 and pooled specificity was 0.94

The same figures appear in the US Preventive Services Task Force's evidence review of 2021 (14 studies, 45,403 people). The advanced-adenoma figures come from a different pooled analysis (31 studies, 120,255 average-risk people, with colonoscopy as the reference standard), and that paper states plainly:

Performance characteristics are for 1-time rather than serial testing.

[Note 8] Positive predictive value against prevalence, from the WHO's Screening programmes: a short guide (2020):

When a condition is less common in a country (low prevalence), the positive predictive value is lower.

The guide gives a worked example, but not of colorectal cancer and not of FIT: the example is diabetes screening, using one-hour post-prandial blood glucose (Somogyi-Nelson method, threshold 160 mg/100 ml), sensitivity 52.9% and specificity 99.4%; in a population of 10,000 in which 150 truly have the disease, it produces 70 false negatives and 60 false positives. From those two figures that is 80 true positives against 60 false positives, a positive predictive value of about 57% — that 57% is worked out here from the document's own two numbers; the document does not compute it. It is cited for the general rule that a lower prevalence lowers the positive predictive value, not to say that FIT behaves this way: a specificity as high as 99.4% is precisely why the predictive value in that example is so high, and FIT's is not that high. That sensitivity and specificity trade off against one another comes from Wilson and Jungner (1968).

[Note 9] The publication with stratified denominators, from the Centre for Health Protection's Non-Communicable Diseases Watch, March 2025:

Among the participants who had submitted stool specimens with analysable results, around 77 000 persons (16%) had positive FIT results.

Among FIT-positive participants who underwent a colonoscopy examination, around 3 400 had colorectal cancer and around 40 000 cases of colorectal adenoma (a type of polyp that can turn into cancer) have been diagnosed through the Programme.

[Note 10] The concluding sentence of COLONPREV's 10-year result, with its condition in the opening half:

On the basis of participation observed in this study, a faecal immunochemical test-based programme was non-inferior to a colonoscopy-based programme for colorectal cancer-related mortality.

[Note 11] The trial design in the original: "one-time colonoscopy in 26,703 subjects with FIT every 2 years in 26,599 subjects". The 26,703 and 26,599 are the baseline numbers in the first-round report; the 26,332 and 26,719 in the body are the intention-to-screen numbers in the ten-year report, counted differently, and the two sets should not be treated as one. The factors of 2.22 and 9.28 are worked out here from the paper's own counts; the trial reports odds ratios of 2.30 and 9.80. Note that dividing the rounded percentages instead gives 2.1 and 10.5, the first too low and the second too high, and that is not how it should be done. The factor of 1.96 is likewise this article's division of two published percentages (39.4 ÷ 20.1 = 1.96); the breakdown is not in that paper's abstract and is taken here at second hand from a commentary, with both sides reconciling to their totals (39.4 + 0.5 = 39.9; 20.1 + 11.7 = 31.8).

[Note 12] The concluding sentence of NordICC's 13-year report:

One colonoscopy significantly reduced colorectal cancer incidence but not mortality over 13 years.

NordICC enrolled people aged 55 to 64 in Poland, Norway, Sweden and the Netherlands, while the 10-year and 13-year reports analyse only Poland, Norway and Sweden, the Netherlands being excluded. Participation across the four countries ran from 22.9% (the Netherlands) to 60.7% (Norway), a reminder that these participation rates are made by systems and are not an intrinsic property of the test.

[Note 13] The three guaiac randomised trials. Minnesota (46,551 people, aged 50 to 80): cumulative 13-year colorectal cancer mortality per thousand of 5.88 in the annually screened group (95% CI 4.61–7.15) against 8.83 in the control (95% CI 7.26–10.40), with the biennial group at 8.33 (95% CI 6.82–9.84) — a 33% reduction with annual screening, while the difference between the biennial group and the control did not reach statistical significance; the paper states "About 83 percent of the slides were rehydrated". Nottingham found a 15% reduction (OR 0.85, 95% CI 0.74–0.98; in absolute terms 360/75,253 = 0.48% against 420/74,998 = 0.56%). Funen found a mortality ratio of 0.82 (95% CI 0.68–0.99; in absolute terms 205/30,967 = 0.66% against 249/30,966 = 0.80%). Those percentages are worked out here from the papers' own counts.

[Note 14] The Expert Working Group's recommendation in the original (that document is published in English only):

For asymptomatic population at average risk, individuals aged 50 to 75 years should consider screening by one of the following screening methods - annual or biennial faecal occult blood test; sigmoidoscopy every 5 years; or colonoscopy every 10 years.

The Centre for Health Protection's Chinese summary table of 2024 carries the official Chinese version of the same recommendation. The recommendation was made in 2010, confirmed as still valid in November 2012, updated in June 2022, and endorsed by the Cancer Coordinating Committee on 28 June 2024. The chain of rules begins with the fourth and seventh of the WHO's ten screening principles of Wilson and Jungner, 1968: that the condition should have a recognisable latent or early symptomatic stage, and that its natural history should be adequately understood.

One frequently cited Legislative Council reply is worth reading for what it actually answers — whether more cancer screening programmes should be launched, not why the colorectal programme sets the boundaries it does:

Excessive screening under public health programme not only wastes resources for the overall public health, but also runs out of resources that can be invested on other projects in greater need, and may pose unnecessary health risks to individuals.

[Note 15] The consequence of a false negative, in the programme's own words:

will be falsely reassured leading to delay in seeking appropriate medical following up or treatment and hence affecting the prognosis

That last clause is the most important part of the sentence. The WHO's guide on the weight of screening's harms:

Because most people who are screened do not have the condition, more people can be exposed to the harm of screening than may be able to benefit from it.

[Note 16] The Government's qualitative statement on colonoscopy complications, from the Cancer Online Resource Hub:

Major complications such as perforation, bleeding, heart and lung complications, infection or acute intestinal obstruction are uncommon, with a risk of less than 1%. Patients should seek immediate advice from doctors if they experience abdominal pain or note blood in stools after the procedure.

The pooled paper's own sentence: "Complication rate was lower for screening/surveillance than for diagnostic examinations." The conversions to 0.31/1,000 and 1.46/1,000, and the factors of 1.6 and 1.8, are all worked out here.

[Note 17] Lead-time bias. The WHO's sentence only makes sense read to the end:

Screening can increase survival time from diagnosis but does not necessarily affect when the person will die from the condition, only that they survive longer with a diagnosis.

The guide names it in the next sentence: "This is called lead-time bias". The other bias is that screening more readily finds slow-growing, less aggressive cancers — the WHO guide calls this length-time bias, while the US National Cancer Institute uses length-biased sampling or length bias (that page states "Updated: July 16, 2018" at its foot; the 2012 shown in the URL cited here is the year of publication, not the last update).

[Note 18] The official explanation for case numbers and rates diverging, giving three causes:

largely driven by the ageing population, along with evolving cancer risks and improvements in diagnostic practices

On the split between age groups, the official key findings read:

Colorectal cancer: Rising in younger adults but rates in older adults in both sexes have been declining since 2017.

The WHO's guide on the trap of before-and-after comparison: the improvement may come from "better diagnosis and treatment in the population rather than the screening programme".

[Note 19] The description of the Australian programme, taken from a web archive capture of the department's page dated 2026-02-18 rather than the live page. The three eligibility conditions: "are between 45 and 74"; "have a green Medicare card or registered as a Department of Veterans' Affairs customer"; "have an Australian mailing address." The same page states "We get your details from Medicare." On posting:

People aged 50 to 74 receive a bowel cancer screening kit in the post every 2 years.

People aged 45 to 49 can request their first bowel cancer screening kit by submitting a webform or calling 1800 627 701.

The same page also states "From 1 July 2024, people aged 45 to 49 can join the program and screen for free."

[Note 20] Row 3 of the first table in the WHO's Screening programmes: a short guide, the dimension being "How is the test offered?" The ad hoc cell:

Participation is often poor, and those that use the services tend to have higher socioeconomic status

The cell for call and recall from a register of the eligible population:

Increases participation and limits inequities linked to socioeconomic level

---

#Sources and dates checked

• The four eligibility conditions and both footnotes verbatim, the birth years 1950–1976, the Electronic Health System requirement, and the doctor's in-person assessment: Colorectal Cancer Screening Programme, "Who may enrol" (Chinese page last revised 07 April 2025) https://www.colonscreen.gov.hk/tc/public/programme/who_may_enrol.html; "Who may enrol" (Last Revision Date: 2025-01-02) https://www.colonscreen.gov.hk/en/public/programme/who_may_enrol.html, retrieved 2026-08-02 (each page retrieved separately; the two are not merged).
• The definition of screening, the six situations that rule a person out and the three that make joining unnecessary, the comparable effectiveness of the three methods, and the false-negative passage: "Understand the suitability for receiving CRC screening" (Chinese page last revised 10 April 2025) https://www.colonscreen.gov.hk/tc/public/programme/understand_the_suitability_for_receiving_crc_screening.html; the English page (Last Revision Date: 2025-4-10 — same day as the Chinese, formatted differently) https://www.colonscreen.gov.hk/en/public/programme/understand_the_suitability_for_receiving_crc_screening.html, retrieved 2026-08-01.
• The five common symptoms and their open lead-in "Common signs and symptoms of colorectal cancer include:", "early colorectal cancer may have no symptoms at all", the sentence that people at high risk and people with common symptoms should consult a doctor and be examined regularly or as soon as possible, and five-year relative survival by stage: Government Cancer Online Resource Hub, "Colorectal Cancer" https://www.cancer.gov.hk/tc/hong_kong_cancer/common_cancers_in_hong_kong/colorectal_cancer.html; English version https://www.cancer.gov.hk/en/hong_kong_cancer/common_cancers_in_hong_kong/colorectal_cancer.html (neither page carries a revision date), retrieved 2026-08-03.
• The $280 subsidy at the FIT stage, the $76 administrative subsidy, the extra fee chargeable on top of the subsidy with no stated cap, and the four things the doctor may not charge for (including obtaining screening advice): "Subsidy level and co-payment" (primary care doctor; Chinese page last revised 6 August 2018) https://www.colonscreen.gov.hk/tc/service/primary_care_doctor/subsidy_level_and_copayment.html; English version (Last Revision Date: 2018-8-6) https://www.colonscreen.gov.hk/en/service/primary_care_doctor/subsidy_level_and_copayment.html, retrieved 2026-08-09 (each retrieved separately; the duplicated character in the Chinese page's 「獲資助的的診症服務」 is the source's own).
• The $8,500/$7,800 colonoscopy subsidy, the $300 of it for the pre-colonoscopy consultation subsidised once per participant, the requirement that subsequent pre-colonoscopy consultations be paid out of pocket, the $1,000 cap and its purpose, and the charity quota: "Subsidy level and co-payment" (Chinese page last revised 6 August 2018) https://www.colonscreen.gov.hk/tc/service/colonoscopist/subsidy_level_and_copayment.html; English version https://www.colonscreen.gov.hk/en/service/colonoscopist/subsidy_level_and_copayment.html, retrieved 2026-08-01.
• FOBT detecting blood in the stool invisible to the eye, and removing polyps to prevent them becoming cancer: "Background of Programme" (Last Revision Date: 2021-12-03) https://www.colonscreen.gov.hk/en/public/programme/background_of_programme.html; the two sample bottles and collection from two separate movements: "How to collect stool specimen" (Last Revision Date: 2020-1-22) https://www.colonscreen.gov.hk/en/public/programme/how_to_collect_stool_specimen.html; return within 4 days: "Returning specimens" https://www.colonscreen.gov.hk/en/public/programme/returning_specimens.html, retrieved 2026-08-01.
• The eligible birth years for 2026, the two-stage pathway, cumulative participation above 579,000 and the release's 15%/7.9%/0.7% (the release does not state the denominator of any of them) and "over 50% early stage", the $1,000 cap, over 70% of service locations charging no extra fee, and the annual update of birth years: Department of Health press release 2025-12-31 (English) https://www.info.gov.hk/gia/general/202512/31/P2025123100231.htm; Chinese version https://www.info.gov.hk/gia/general/202512/31/P2025123100227.htm, retrieved 2026-08-01.
• The history of the CEWG recommendation (2010/2012/2022/2024), the three methods for average risk at 50–75, the programme's course from the 2016 pilot to full rollout in 2020, and the reasons for not lowering the age to 40: CEWG, Recommendations on Prevention and Screening for Colorectal Cancer, For Health Professionals (June 2024) https://www.chp.gov.hk/files/pdf/cewg_crc_professional_hp.pdf; the 2016 version of the recommendation (March 2017) https://www.chp.gov.hk/files/pdf/cewg_crc_professional.pdf, retrieved 2026-08-01.
• The official Chinese version of the CEWG recommendation: summary tables of the Cancer Expert Working Group's recommendations (2024 edition) https://www.chp.gov.hk/files/pdf/13_summary_tables_on_cewg_recommendations_on_crc_cervical_and_breast-cancer_chi.pdf, retrieved 2026-08-01.
• The CEWG's use of the Wilson and Jungner principles as its guiding principles: Recommendations of Cancer Expert Working Group on Cancer Prevention and Screening – An Overview for Health Professionals (June 2018) https://www.chp.gov.hk/files/pdf/overview_of_cewg_recommendations_professional_hp.pdf, retrieved 2026-08-01.
• Excessive screening wasting resources and posing health risks: Legislative Council LCQ19, "Prevention of cancers", 2024-05-22 https://www.info.gov.hk/gia/general/202405/22/P2024052200498.htm, retrieved 2026-08-01.
• The fourth and seventh of the ten screening principles, the trade-off between sensitivity and specificity, and the low-prevalence false-positive worked example: Wilson JMG, Jungner G. Principles and practice of screening for disease. WHO Public Health Papers 34. Geneva: WHO; 1968 https://iris.who.int/items/7a67638b-14e3-4dba-b013-dc206ff279d8, retrieved 2026-08-01.
• Positive predictive value falling with prevalence, the structural reason screening does harm, further examination required for false positives, the definition of overdiagnosis, lead-time bias, row 3 of Table 1 "Dimensions of screening", and the rule against using before-and-after mortality comparison as evidence: WHO Regional Office for Europe. Screening programmes: a short guide. Copenhagen; 2020 https://iris.who.int/items/ba910aa7-3c5a-4cf4-97d8-b128ac712881, retrieved 2026-08-01.
• The definition of length-time bias: National Cancer Institute, Crunching Numbers: What Cancer Screening Statistics Really Tell Us (2012-05-01) https://www.cancer.gov/about-cancer/screening/research/what-screening-statistics-mean, retrieved 2026-08-01.
• COLONPREV's 10-year mortality, "any screening at all" of 31.8% against 39.9%, RR 0.92 (0.64–1.32), the non-inferiority margin, the conditional clause "On the basis of participation observed in this study", and the enrolment age range 50–69: Castells A, Quintero E, Bujanda L, et al. Lancet 2025;405(10486):1231-1239 (PMID 40158525; DOI 10.1016/S0140-6736(25)00145-X). The participation breakdown (FIT arm 39.4% FIT / 0.5% colonoscopy; colonoscopy arm 20.1% colonoscopy / 11.7% FIT) and NordICC's adherence by country (Norway 60.7%, Sweden 39.8%, Poland 33.0%, the Netherlands 22.9%): Brenner H, Hoffmeister M. Effective Colorectal Cancer Screening—Adherence is Key. United European Gastroenterol J 2026;14(1):e70180 (PMID 41579066; PMC12831479, open access), retrieved 2026-08-09. That article's PubMed publication type is Editorial — a commentary on COLONPREV, not a report of the trial itself; the participation breakdown and the country adherence rates above are taken at second hand from it, not from the trial report. The design of a one-time colonoscopy against a FIT every two years, the first-round adenoma and cancer counts and the first-round examination rates (24.6%/34.2%), and the enrolment age range 50–69: Quintero E, Castells A, Bujanda L, et al. N Engl J Med 2012;366(8):697-706 (PMID 22356323). The adherence-adjusted post-hoc analysis: Castells A, Quintero E, Chen H, et al. Gut 2026 (PMID 42270316).
• NordICC's 42.0% participation, the 10-year incidence and mortality risks, the 15 major bleeds after polypectomy, the enrolment age range 55–64 and the four-country enrolment against three-country analysis: Bretthauer M, Løberg M, Wieszczy P, et al. N Engl J Med 2022;387(17):1547-1556 (PMID 36214590). The 13-year incidence 1.46% against 1.80% (RR 0.81, 0.71–0.90), mortality 0.41% against 0.47% (RR 0.88, 0.68–1.08), per-protocol RR 0.55 (0.33–0.81) and 0.70 (0.26–1.25), and the concluding sentence: Kaminski MF, Kalager M, Løberg M, et al; NordICC Study Group. Lancet 2026;407(10541):1787-1795 (PMID 42102826; DOI 10.1016/S0140-6736(26)00508-8), retrieved 2026-08-09.
• The three-arm randomised trial of one-off colonoscopy, two rounds of FIT and unscreened usual care (278,280 people, aged 60, median follow-up 4.8 years): Westerberg M, Ludvigsson JF, Metcalfe C, et al. Colonoscopy and fecal immunochemical testing versus usual care in diagnostic colorectal cancer screening: the SCREESCO randomized controlled trial. Nat Med 2026;32(4):1278-1285 (PMID 41721101; DOI 10.1038/s41591-026-04225-9), retrieved 2026-08-09.
• FIT's pooled sensitivity 0.74 and specificity 0.94 for colorectal cancer at 20 µg/g: paragraph 14 of the CEWG professional document of June 2024 (above); the same figures also appear in Lin JS, Perdue LA, Henrikson NB, et al. JAMA 2021;325(19):1978-1998 (PMID 34003220). FIT's single-test sensitivity for advanced adenoma of 0.40 (95% CI 0.33–0.47) and the limitation sentence "1-time rather than serial testing": Imperiale TF, Gruber RN, Stump TE, Emmett TW, Monahan PO. Ann Intern Med 2019;170(5):319-329 (PMID 30802902). Screening-specific colonoscopy complications (perforation 3.1/10,000, major bleeding 14.6/10,000): Lin JS, et al. JAMA 2021;325(19):1978-1998 (as above), retrieved 2026-08-09.
• The faecal occult blood mortality trials, with slide rehydration (83%), the volunteer population, Funen's incidence of 481 against 483, and the absolute death counts in each trial: Mandel JS, et al. N Engl J Med 1993;328(19):1365-71 (PMID 8474513); Hardcastle JD, et al. Lancet 1996;348(9040):1472-7 (PMID 8942775); Kronborg O, et al. Lancet 1996;348(9040):1467-71 (PMID 8942774).
• The pooled colonoscopy complication rates (all indications, including the sentence "Complication rate was lower for screening/surveillance than for diagnostic examinations"): Reumkens A, et al. Am J Gastroenterol 2016;111(8):1092-101 (PMID 27296945). The Hong Kong Government's "the risk of major complications is less than 1%" and its instruction to seek care after the procedure, and the official Chinese version of the CEWG's advice for higher-risk people (Lynch syndrome from age 25, familial adenomatous polyposis from age 12, the family-history threshold and the 10-years-earlier rule): Government Cancer Online Resource Hub, "Colorectal Cancer" (above), retrieved 2026-08-09.
• Colorectal cancer ASR, case numbers, and the ten-year average annual percentage change: Colorectal Cancer in 2023 (August 2025) https://www3.ha.org.hk/cancereg/pdf/factsheet/2023/colorectum_2023.pdf; Colorectal Cancer in 2021 (October 2023) https://www3.ha.org.hk/cancereg/pdf/factsheet/2021/colorectum_2021.pdf; Colorectal Cancer in 2019 (October 2021) https://www3.ha.org.hk/cancereg/pdf/factsheet/2019/colorectum_2019.pdf, retrieved 2026-08-01.
• The 5,467 new cases in 2023 and the rankings, the 2,266 colorectal cancer deaths being 15.2% of cancer deaths, the 2013-against-2023 ten-year comparison, all new cancers growing 2.7% a year against population growth of 0.5%/4.9%, the population-ageing explanation, the age-specific trends in Figure 7, the sentence "rates in older adults in both sexes have been declining since 2017", and the caution that a trend needs five years or more: Overview of Hong Kong Cancer Statistics of 2023 (August 2025) https://www3.ha.org.hk/cancereg/pdf/overview/Overview%20of%20HK%20Cancer%20Stat%202023.pdf, retrieved 2026-08-01.
• The ever-screened proportions at 50–75 of 33.2%/42.6%/46.2% (the data labels of the first figure; 33.2% and 42.6% from the Population Health Survey, 46.2% from the 2023 Behavioural Risk Factor Survey, not the same instrument) and the statement that the screening rate still needs improvement: Centre for Health Protection, Non-Communicable Diseases Watch, March 2025: Colorectal Cancer Screening and Prevention (the English edition of the same issue) https://www.chp.gov.hk/files/pdf/ncd_watch_mar_2025_en.pdf, retrieved 2026-08-01.
• The defining sentence "around 56% were at an early stage (i.e. stage II or below)": Centre for Health Protection, Non-Communicable Diseases Watch, March 2025 (Chinese edition, below), retrieved 2026-08-09; the same definition in English (stage II or below, about 57%): Legislative Council LCQ19 (above). The 2023 territory-wide colorectal cancer stage distribution (stage I 9.8%, stage II 19.4%, stage III 26.6%, stage IV 23.9%, unstaged 20.4%): Colorectal Cancer in 2023 (above), retrieved 2026-08-09.
• Australia posting the iFOBT kit to eligible people's Australian mailing addresses, automatic posting every two years at 50 to 74, the requirement that people aged 45 to 49 request their first kit, the lowering of the eligible age to 45 from 1 July 2024, and the eligible range of 45 to 74: About the National Bowel Cancer Screening Program, Australian Government Department of Health, Disability and Ageing https://www.health.gov.au/our-work/national-bowel-cancer-screening-program/about-the-national-bowel-cancer-screening-program (the description quoted here is taken from an Internet Archive capture of that official page dated 2026-02-18, not the live page).
• Australia's 2018–2019 invited-participant rate of 43.5% (2,497,317/5,744,244, ages 50 to 74), the 42.4% for the rolling two-year period 2017–2018, and the fact that participation is computed over rolling two-year periods with adjacent periods overlapping by a calendar year: Australian Institute of Health and Welfare. National Bowel Cancer Screening Program: monitoring report 2021 (AIHW cat. no. CAN 139), 124 pages. The live aihw.gov.au returned HTTP 403 on 2026-08-09, so what was read is the Internet Archive's capture of 2024-06-15, which returned HTTP 200 (1,476,951 bytes), retrieved 2026-08-09.
• The stratified denominators for screening results (about 77 000 people, about 16%, positive among those returning an analysable specimen; about 3 400 colorectal cancers and about 40 000 colorectal adenomas among positives who underwent colonoscopy), the definition of colorectal adenoma, and the ever-screened proportion at 50–75: Centre for Health Protection, Non-Communicable Diseases Watch, March 2025, "Colorectal Cancer Screening and Prevention" (Chinese edition) https://www.chp.gov.hk/files/pdf/ncd_watch_mar_2025_tc.pdf, retrieved 2026-08-02.

This article was written from the sources listed above. It is about eligibility, the structure of the charges and the evidence; whether you are suitable to join, it does not answer — that is decided by a primary care doctor at the consultation.

---

#Further reading

• On this platform: the article on gastroscopy and colonoscopy charges
• Colorectal Cancer Screening Programme: colonscreen.gov.hk